Germline transformationof the butterflyBicyclus anynanaJeffrey M. Marcus†, Diane M. Ramosand Antonia Monteiro*

Department of Biological Sciences, University at Buffalo, Buffalo,NY 14260, USA* Author for correspondence (monteiro@buffalo.edu).

Recd 16.12.03; Accptd 26.01.04; Published online 10.03.04

Ecological and evolutionary theory has frequentlybeen inspired by the diversity of colour patterns onthe wings of butterflies. More recently, these variedpatterns have also become model systems for study-ing the evolution of developmental mechanisms. Atechnique that will facilitate our understanding ofbutterfly colour-pattern development is germlinetransformation. Germline transformation permitsfunctional tests of candidate gene products and ofcis-regulatory regions, and provides a means of gen-erating new colour-pattern mutants by insertionalmutagenesis. We report the successful transform-ation of the African satyrid butterfly Bicyclusanynana with two different transposable elementvectors, Hermes and piggyBac, each carrying EGFPcoding sequences driven by the 3XP3 syntheticenhancer that drives gene expression in the eyes.Candidate lines identified by screening for EGFP inadult eyes were later confirmed by PCRamplification of a fragment of the EGFP codingsequence from genomic DNA. Flanking DNA sur-rounding the insertions was amplified by inversePCR and sequenced. Transformation rates were 5%for piggyBac and 10.2% for Hermes. Ultimately, thenew data generated by these techniques may permitan integrated understanding of the developmentalgenetics of colour-pattern formation and of the eco-logical and evolutionary processes in which thesepatterns play a role.

Keywords: germline transformation; butterfly; genetics;transposons; evolution and development;Bicyclus anynana

1. INTRODUCTIONBiological hypotheses ranging from the ecological role ofaposematic coloration (Wallace 1881; Nishida 2002) andthe evolution of mimicry (Bates 1863; Kapan 2001), tothe basis of phenotypic plasticity (Merrifield 1892;Brakefield et al. 1998) have historically been inspired bythe eye-catching colour patterns on the wings of butter-flies. More recently, butterfly colour patterns have alsobecome an important model system for understanding therelationship between development and evolution because

† Present address: Department of Biology, Western Kentucky University,Bowling Green, KY 42101, USA.

Proc. R. Soc. Lond. B (Suppl.) 271, S263–S265 (2004) S263 2004 The Royal SocietyDOI 10.1098/rsbl.2004.0175

they are highly variable, consist of clearly defined sub-units, exist in two dimensions and yet are structurally verysimple, making them very amenable for study andmanipulation (Nijhout 1991; Beldade & Brakefield 2002;McMillan et al. 2002).

Unfortunately, progress in understanding the develop-mental genetic processes underlying butterfly colour-pattern formation has been limited, for two reasons. First,researchers have not been able to characterize any of themutations that alter colour patterns (Weatherbee et al.1999; Monteiro et al. 2003), so there is little mechanisticunderstanding of how mutant colour-pattern phenotypesare produced. Second, while gene expression patterns thatresemble adult colour patterns are suggestive (Carroll etal. 1994; Brunetti et al. 2001), there are very few dataavailable showing that these gene products have a func-tional role in colour-pattern formation. Germline trans-formation is one genetic technology that has been used forboth in vivo tests of gene function (Brand & Perrimon1993) and for the production of mutations that are easyto characterize at the molecular level (Cooley et al. 1988).

Inspired by the recent studies that identify several trans-posable elements capable of insertion into multiple arthro-pod genomes (Berghammer et al. 1999), including twospecies of moths (Peloquin et al. 2000; Tamura et al.2000), we tested whether two transposons, Hermes andpiggyBac, are capable of inserting into the genome of theAfrican satyrid butterfly Bicyclus anynana (figure 1a). Thesuccess of these experiments represents the first demon-stration, to our knowledge, of germline transformation ina butterfly.

2. MATERIAL AND METHODSWe tested whether the piggyBac construct pBac[3xP3-EGFP] and

the Hermes construct Her[3xP3-EGFP] (generously provided by ErnstWimmer), carrying the marker gene for enhanced green fluorescentprotein (EGFP), could insert into the germline of the butterfly B.anynana. Eggs were collected from maize leaves following 1 h ovi-position bouts, and placed into glass Petri dishes on thin strips ofdouble-sided adhesive tape. Eggs were injected with equal concen-trations (500 ng µl�1) of one of the plasmids mentioned above, and ahelper plasmid containing the coding sequence of either the piggyBac(plasmid construct pHsp82Pbac) or Hermes transposase (plasmidconstruct pKhsp82Hermes) driven by the Drosophila heat-shock pro-moter (Horn et al. 2000), using a pulled glass needle attached to aPicospritzer III microinjection apparatus. After injection, embryoswere placed in an incubator at 27 °C and 80% relative humidity untilthe larvae hatched 4–6 days later, and were transferred to food plantswith a small paintbrush. Adults reared from injected eggs were matedin individual cages with 3–5 virgin individuals of the opposite sex toestablish families. Eggs were collected from each family, and the lar-vae were screened for the presence of EGFP in their six larval stem-mata, and again in the adult compound eye with a Nikon SMZ1500fluorescent microscope.

All EGFP-positive F1 individuals were crossed separately withwild-type, whereas in all subsequent generations, matings were con-ducted within families to generate homozygous lines. We confirmedthe presence of the EGFP gene in all of these families by PCR(forward primer EGFPFlas CGT GAC CAC CCT GAC CTAC,reverse primer EGFPRlas TGA TCG CGC TTC TCG TT, PCRconditions: 1 × 94 °C, 2 min; 40 × (94 °C, 30 s; 58.2 °C, 30 s; 72 °C,1 min); 1 × 72 °C, 6 min). The transposon insertion sites in transfor-med families were amplified by inverse PCR (genomic DNA digestedby HaeIII, MspI, Sau3a or TaqI restriction enzymes, circularized byT4 ligase and amplified with primer pairs PLF and PLR, PRF andPRR, HLF and HLR, and HRF and HRR (Horn & Wimmer 2000),PCR conditions: 1 × 95 °C, 5 min; 30 × (95 °C, 30 s; 65 °C, 1 min;68 °C, 2 min); 1 × 72 °C, 10 min; some PCR products were reampli-fied with the same primers and conditions to increase productconcentration) and sequenced. Flanking sequences were then exam-ined to determine whether the sequences contained the expectedtarget-site duplications for each transposon.

S264 J. M. Marcus and others Butterfly germline transformation

F6P

F20P

F5H

F16H

M13H–1

M13H–2

M21H

ND

GTCTTGTAAGAACCTCTTAA

ND

ND

ND

ND

GCTATGACCGTCGTGAGTAC

Piggybac TTAATACATTGCTAGCTAGT

Piggybac TTAACAAAGTTAATAAAATG

Hermes GTGGACTCCAGTCTCTCGTC

Hermes GTGCAGGTATTTTAGGCGCA

Hermes GTGACGCACAAAATACTATG

Hermes ATGGCAGCCCACATCCCGTC

Hermes GTGGCTGCATTCACGCACAG

750500

bpF6P F20P F5H F15H M13H

(1)M13H(2)

M21H Wt

(a)(i)

(b) (c) (d )

(e) ( f ) (g)

(h)

(i)

(ii)

Figure 1. Germline transformation of the butterfly Bicyclusanynana. (a) (i) Ventral and (ii) dorsal views of wild-type B.anynana. (b–d) Adult B. anynana eyes under bright-fieldillumination. (e–g) The same butterflies under epifluorescentillumination. (b,e) Wild-type B. anynana. (c,f ) Bicyclusanynana transformed with piggyBac [3XP3-EGFP]. (d,g)Bicyclus anynana transformed with Hermes [3XP3-EGFP].(h) PCR amplification of EGFP inserts in transformed linesof B. anynana. (i) Partial flanking sequences surroundingtransposon insertion sites. Proposed piggyBac and Hermestarget site duplications are shown in bold. In one case,M13H, different flanking sequences were obtained fromdifferent sublines descended from a single injectedindividual. Both sublines test positive for EGFP expressionboth by visual inspection and by EGFP PCR amplification(see (h)). We suggest that the original injected individualthat founded this subline had more than one Hermesinsertion, most probably on different chromosomes, andthese insertions were then separated from one another bygenetic segregation when mated to wild-type animals in theF1 generation.

3. RESULTSIn total, 3357 eggs were injected with the piggyBac con-

struct and 6741 eggs were injected with the Hermes con-struct. Approximately 95% of the injected embryos diedbefore hatching, presumably as a result of dehydration viathe punctured hole in the chorion. Control (uninjected)

Proc. R. Soc. Lond. B (Suppl.)

eggs that were otherwise treated in the same way asinjected eggs have a hatch rate of ca. 80%. Out of theseinjected eggs, 40 animals derived from eggs injected withpiggyBac and 39 animals derived from eggs injected withHermes survived to adulthood and produced offspringwhen mated to wild-type uninjected individuals. From thefamilies derived from the injected individuals, we ident-ified two piggyBac families (5.0%) and four Hermes famil-ies (10.2%) with one or more offspring expressing EGFP(figure 1b–g). Offspring from these EGFP-expressing indi-viduals backcrossed to wild-type, produced PCR amplifi-cation products when amplified with EGFP primers(figure 1h). Finally, we used inverse PCR to sequenceflanking regions of each insertion. We were not able tosequence both ends of every insertion, but all of thesequences that we were able to obtain included theexpected target-site duplications (figure 1i). We con-tinued rearing and inbreeding three lines of transgenicbutterflies (one piggyBac and two Hermes). The presenceof the EGFP gene was again confirmed by PCR in individ-uals of the fifth generation.

4. DISCUSSIONOur findings demonstrate that transposon-based germ-

line transformation in butterflies is possible and takesplace at efficiencies comparable to those reported in mothtransformation experiments (Peloquin et al. 2000; Tamuraet al. 2000). Although the transformation rate amonginjected individuals that survived injection is reasonablygood, improvements in the injection protocol to increasesurvivorship are needed to increase the efficiency ofexperiments such as those described here, and this is asubject that is currently under investigation. The detectionof EGFP-positive individuals was often difficult owing tothe faint EGFP expression in most of the transformedindividuals. This may have been caused by position effectson gene expression or the presence of the normal eye pig-ments that could block the EGFP signal. Detection ofEGFP-positive individuals can be facilitated by digitalphotographic techniques, and our ability to detect trans-formants with weak EGFP expression might have beengreater had we used digital imaging as part of the screen-ing process. Our transformation rates should therefore beviewed as minimum estimates. Finally, it appears thatmale B. anynana have slightly darker eyes than females,so we recommend screening the sexes separately.

With slight modifications to the injected constructs, thetechnique of germline transformation can be used forenhancer trapping and insertional mutagenesis screens formutations that alter colour-pattern phenotypes. Further,misexpression constructs of candidate genes or reporterconstructs with candidate cis-regulatory regions can beincluded within a transposon vector and introduced intothe butterfly germline so that their role in colour-patternformation can be tested in vivo. This ability to manipulatea butterfly genome will greatly advance our understandingof how butterfly colour patterns develop and allow us tointegrate fully the development of these patterns with theirecology and evolution.

AcknowledgementsThe authors thank E. Wimmer for plasmid constructs, D. Pietras forplant husbandry and W. Piel and two anonymous referees for

Butterfly germline transformation J. M. Marcus and others S265

comments on the manuscript. J.M.M. received postdoctoral supportfrom SUNY at Buffalo. D.M.R. was supported by a NSF IGERTbiophotonics predoctoral traineeship. This work was supported by agrant from the NSF (award 29339) to A.M.

Bates, H. W. 1863 The naturalist on the river Amazon. London:John Murray.

Beldade, P. & Brakefield, P. M. 2002 The genetics and evo–devo ofbutterfly wing patterns. Nat. Rev. Genet. 3, 442–452.

Berghammer, A. J., Klingler, M. & Wimmer, E. A. 1999 Genetictechniques—a universal marker for transgenic insects. Nature 402,370–371.

Brakefield, P. M., Kesbeke, F. & Koch, P. B. 1998 The regulationof phenotypic plasticity of eyespots in the butterfly Bicyclusanynana. Am. Nat. 152, 853–860.

Brand, A. H. & Perrimon, N. 1993 Targeted gene expression as ameans of altering cell fates and generating dominant phenotypes.Development 118, 401–415.

Brunetti, C. R., Selegue, J. E., Monteiro, A., French, V., Brakefield,P. M. & Carroll, S. B. 2001 The generation and diversification ofbutterfly eyespot colour patterns. Curr. Biol. 11, 1578–1585.

Carroll, S. B., Gates, J., Keys, D. N., Paddock, S. W., Panganiban,G. E. F., Selegue, J. E. & Williams, J. A. 1994 Pattern formationand eyespot determination in butterfly wings. Science 265, 109–114.

Cooley, L., Kelley, R. & Spradling, A. C. 1988 Insertionalmutagenesis of the Drosophila genome with single P-elements.Science 239, 1121–1128.

Horn, C. & Wimmer, E. A. 2000 A versatile vector set for animaltransgenesis. Dev. Genes Evol. 210, 630–637.

Proc. R. Soc. Lond. B (Suppl.)

Horn, C., Jaunich, B. & Wimmer, E. A. 2000 Highly sensitive, fluor-escent transformation marker for Drosophila transgenesis. Dev.Genes Evol. 210, 623–629.

Kapan, D. D. 2001 Three-butterfly system provides a field test ofmullerian mimicry. Nature 409, 338–340.

McMillan, W. O., Monteiro, A. & Kapan, D. D. 2002 Developmentand evolution on the wing. Trends Ecol. Evol. 17, 125–133.

Merrifield, F. 1892 Conspicuous effects on the markings and colour-ing of Lepidoptera caused by exposure of the pupae to differenttemperature conditions. Trans. Entomol. Soc. Lond. 1892, 33–44.

Monteiro, A., Prijs, J., Hakkaart, T., Bax, M. & Brakefield, P. M.2003 Mutants highlight the modular control of butterfly eyespotpatterns. Evol. Dev. 5, 180–187.

Nijhout, H. F. 1991 The development and evolution of butterfly wingpatterns. Washington, DC: Smithsonian Institution Press.

Nishida, R. 2002 Sequestration of defensive substances from plantsby Lepidoptera. A. Rev. Entomol. 47, 57–92.

Peloquin, J. J., Thibault, S. T., Staten, R. & Miller, T. A. 2000Germ-line transformation of pink bollworm (Lepidoptera:Gelechiidae) mediated by the piggyBac transposable element. InsectMol. Biol. 9, 323–333.

Tamura, T. (and 13 others) 2000 Germline transformation of thesilkworm Bombyx mori L. using a piggyBac transposon-derived vec-tor. Nat. Biotech. 18, 81–84.

Wallace, A. R. 1881 Protective mimicry in animals. In Science for all(ed. R. Brown), pp. 138–147. London: Cassell, Petter, Galpin &Co.

Weatherbee, S. D., Nijhout, H. F., Grunert, L. W., Halder, G.,Galant, R., Selegue, J. & Carroll, S. 1999 Ultrabithorax functionin butterfly wings and the evolution of insect wing patterns. Curr.Biol. 9, 109–115.